재무이익-세무이익간의 일시적 차이 구성항목의 내용과 지속성

초록

The difference between financial accounting income and taxable income is commonly referred to as the book-tax differences. Researchers used this measure as a proxy for earnings quality or as a proxy for level of corporate tax planning. BTD researches usually use the total amount of BTDs. In drawing policy implications from observed BTDs, an understanding of the contents of BTDs and their properties is important. In particular, identifying the degree of persistence of BTDs is critical. The main purpose of this paper is to investigate the contents and persistence of BTDs and their components. We cannot obtain information on permanent differences form audit reports, so focus on temporary differences(TDs), and analyse the contents and persistence of TDs. In our analysis, we use Korean listed firms with available Korea Investors Services (KIS)-Value data for 15 years from 1999 to 2013. We restrict the sample to non-financial firms and have a december year-end. Additionally, firm-years with missing financial data which can not be used in persistence analysis are excluded. We delete observations in which the absolute value of TDs exceeds 0.3. These data requirements result in a sample of 7,298 firm-year observations for analysing the contents of TDs and 6,142 firm-year observations for analysing the persistence of TDs. We hand-collected temporary differences information from the tax notes in the audit report for 15 years from 1999 to 2013. Initially 41 detailed items of TDs are classified, and we analyse the contents and changes of these items by year and by industry. Next, we grouped these items into 13 components considering the size of items and analyse the persistence of these components. To analyse the persistences of TDs, ordinary least square method is used in main analysis. Additional analysis using quantile regression is used to describe the more detailed relationship at different points in the conditional distribution of dependent variable. The findings in this paper are as follows. First, the amount of inclusion in deductible expenses, and that of inclusion in gross income does not change so much form 1999 to 2006. But the amounts increase from 2007 to 2010, and the size of TDs is also increased in this period. Second, large number of components affects the change of sign and size of TDs, and the change are originated from various sources, for example, the introduction of IFRS, accounting standard adjustment around financial crisis, and the change of tax rules. Third, the average consecutive years of same sign of TDs are estimated to 1.32 years. This means the persistence of TD sign is relatively low and TDs are reversed at short time period. Fourth, the persistence of TD for 15 years is estimated to 0.101. But there is wide variation in persistence by year and by industry, and the accrual and reversal of TD components and persistence of each TD component seem to affect the results. Fifth, the result of quantile regression shows that the persistence of each TD component does not the same at different points in dependent variable.

키워드